tag:blogger.com,1999:blog-17232051.post7931033385118580086..comments2024-03-18T22:32:52.802-04:00Comments on Noahpinion: Equilibria, unique and not-so-unique (guest post by Roger Farmer)Noah Smithhttp://www.blogger.com/profile/09093917601641588575noreply@blogger.comBlogger54125tag:blogger.com,1999:blog-17232051.post-72267971250952930282012-07-16T18:19:12.367-04:002012-07-16T18:19:12.367-04:00Professor Lars Pålsson Syll writes in "The no...Professor Lars Pålsson Syll writes in "The nodal point of the macroeconomics debate"<br />16 July, 2012 <br />http://larspsyll.wordpress.com/2012/07/16/the-nodal-point-of-the-macroeconomics-debate/<br /><br />"This summer both Oxford professor Simon Wren-Lewis and Nobel laureate Paul Krugman have had interesting posts up discussing modern macroeconomics and its alleged needs of microfoundations.<br /><br />Most “modern” mainstream neoclassical macroeonomists more or less subscribe to the view that microfoundations somehow has lead to better models enabling us to make better predictions of future macroeconomic events...Yours truly basically side with Wren-Lewis and Krugman on this issue, but I will try to explain why one might be even more critical and doubtful than they are re microfoundations of macroeconomics.<br /><br />Microfoundations today means more than anything else that you try to build macroeconomic models assuming “rational expectations” and hyperrational “representative actors” optimizing over time. Both are highly questionable assumptions.<br /><br />The concept of rational expectations was first developed by John Muth (1961) and later applied to macroeconomics by Robert Lucas (1972). Those macroeconomic models building on rational expectations-microfoundations that are used today among both “new classical” and “new keynesian” macroconomists, basically assume that people on the average hold expectations that will be fulfilled. This makes the economist’s analysis enormously simplistic, since it means that the model used by the economist is the same as the one people use to make decisions and forecasts of the future.<br /><br />Macroeconomic models building on rational expectations-microfoundations assume that people, on average, have the same expectations. Someone like Keynes for example, on the other hand, would argue that people often have different expectations and information, which constitutes the basic rational behind macroeconomic needs of coordination. Something that is rather swept under the rug by the extremely simple-mindedness of assuming rational expectations in representative actors models, which is so in vogue in “New Classical” and “New Keynesian” macroconomics. But if all actors are alike, why do they transact? Who do they transact with? The very reason for markets and exchange seems to slip away with the sister assumptions of representative actors and rational expectations.<br /><br />Macroeconomic models building on rational expectations microfoundations impute beliefs to the agents that is not based on any real informational considerations, but simply stipulated to make the models mathematically-statistically tractable. Of course you can make assumptions based on tractability, but then you do also have to take into account the necessary trade-off in terms of the ability to make relevant and valid statements on the intended target system. Mathematical tractability cannot be the ultimate arbiter in science when it comes to modeling real world target systems. One could perhaps accept macroeconomic models building on rational expectations-microfoundations if they had produced lots of verified predictions and good explanations. But they have done nothing of the kind"Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-30540304254497046702012-07-16T18:17:35.493-04:002012-07-16T18:17:35.493-04:00Professor Lars Pålsson Syll writes in "The no...Professor Lars Pålsson Syll writes in "The nodal point of the macroeconomics debate"<br />16 July, 2012 <br />http://larspsyll.wordpress.com/2012/07/16/the-nodal-point-of-the-macroeconomics-debate/<br /><br />"This summer both Oxford professor Simon Wren-Lewis and Nobel laureate Paul Krugman have had interesting posts up discussing modern macroeconomics and its alleged needs of microfoundations.<br /><br />Most “modern” mainstream neoclassical macroeonomists more or less subscribe to the view that microfoundations somehow has lead to better models enabling us to make better predictions of future macroeconomic events...Yours truly basically side with Wren-Lewis and Krugman on this issue, but I will try to explain why one might be even more critical and doubtful than they are re microfoundations of macroeconomics.<br /><br />Microfoundations today means more than anything else that you try to build macroeconomic models assuming “rational expectations” and hyperrational “representative actors” optimizing over time. Both are highly questionable assumptions.<br /><br />The concept of rational expectations was first developed by John Muth (1961) and later applied to macroeconomics by Robert Lucas (1972). Those macroeconomic models building on rational expectations-microfoundations that are used today among both “new classical” and “new keynesian” macroconomists, basically assume that people on the average hold expectations that will be fulfilled. This makes the economist’s analysis enormously simplistic, since it means that the model used by the economist is the same as the one people use to make decisions and forecasts of the future.<br /><br />Macroeconomic models building on rational expectations-microfoundations assume that people, on average, have the same expectations. Someone like Keynes for example, on the other hand, would argue that people often have different expectations and information, which constitutes the basic rational behind macroeconomic needs of coordination. Something that is rather swept under the rug by the extremely simple-mindedness of assuming rational expectations in representative actors models, which is so in vogue in “New Classical” and “New Keynesian” macroconomics. But if all actors are alike, why do they transact? Who do they transact with? The very reason for markets and exchange seems to slip away with the sister assumptions of representative actors and rational expectations.<br /><br />Macroeconomic models building on rational expectations microfoundations impute beliefs to the agents that is not based on any real informational considerations, but simply stipulated to make the models mathematically-statistically tractable. Of course you can make assumptions based on tractability, but then you do also have to take into account the necessary trade-off in terms of the ability to make relevant and valid statements on the intended target system. Mathematical tractability cannot be the ultimate arbiter in science when it comes to modeling real world target systems. One could perhaps accept macroeconomic models building on rational expectations-microfoundations if they had produced lots of verified predictions and good explanations. But they have done nothing of the kind"Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-23179776050759265112012-04-23T10:38:27.894-04:002012-04-23T10:38:27.894-04:00P.S.
My own view is that economics is a special ca...P.S.<br />My own view is that economics is a special case of ecology, and it should methods borrowed and adapted from ecology. (i.e. We should be modelling births, deaths and reproduction of interacting and evolving classes of agents.)reasonhttps://www.blogger.com/profile/10958786975015285323noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-89554413579013959222012-04-23T10:33:41.119-04:002012-04-23T10:33:41.119-04:00I think perhaps I'm missing something here. Bu...I think perhaps I'm missing something here. But to be honest, I think this discussion is missing the point by miles. We have just been through a long period of "stability" where there were large persistant current account imbalances, large increases in debt levels and large increases in asset price/income ratios. This "stability" was very unstable and many people noticed and said so. Is this "equilibrium"? If it is, it is a very strange equilibrium, that includes the seeds of its own demise. If the real world doesn't actually follow an equilibrium path, why are why trying to pretend that we can sensibly model it as though it does?reasonhttps://www.blogger.com/profile/10958786975015285323noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-25665729346808807012012-04-21T06:25:42.882-04:002012-04-21T06:25:42.882-04:00Roger,
Thank you for the reply. But mind you, I s...Roger,<br /><br />Thank you for the reply. But mind you, I said that "the multiplicity of equilibria is often a very local phenomenon", and a counter-example is not enough. Generating sunspots commonly means that the researcher must rig the model to get enough roots inside the unit circle. And in a nonlinear model these roots will change value depending on the state of the economy (this is not very precise, but I think you get the idea). I am not, and very few people in the literature are, convinced that this is more than a mathematical curiosity ...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-79773433990748637182012-04-19T23:48:19.171-04:002012-04-19T23:48:19.171-04:00Makes sense to me...Makes sense to me...Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-43208463124527012362012-04-19T22:15:02.627-04:002012-04-19T22:15:02.627-04:00Roger, thanks for your thoughtful reply.
I take i...Roger, thanks for your thoughtful reply.<br /><br />I take it for granted that, taken literally, the assumption of consistent plans is false. Whether the general equilibrium approach is useful must depend on 'how false' this assumption is, i.e. how quickly people adjust their inconsistent expectations so as to become consistent. It doesn't only depend on whether a general equilibrium model can explain the behavior of the variables we care about. (I'm not sure if this is what you meant by "does it help us to understand data?".) Even if we could explain the aggregate variables of interest (during the Depression, say) using a DSGE model, the model would probably make poor out-of-sample predictions unless the assumptions underlying DSGE modelling (rational expectations, market clearing) were close to reality. <br /><br />I don't claim that a viable disequilibrium modelling framework exists (there are many relevant lines of research, but as far as I know most of them were abandoned in the 1970s, and all of them have unresolved problems). However, I don't think it's obvious that a non-DSGE model cannot explain, e.g., persistent unemployment. To explain 15% unemployment lasting, say, 5 years, it's not necessary to assume that wages would take 5 years to adjust to equate labor supply and demand. A key insight of the Barro-Grossman/Dreze/Benassy 'non-Walrasian equilibrium' literature is that disequilibrium in one market can explain unemployment in another market. Somewhat relatedly, there is the old Keynesian argument that even if prices fall when there is excess supply of goods, and nominal wages fall when there is excess supply of labor, *real* wages may not adjust to clear the labor market. (All we get is deflation - which either restores the economy to equilibrium via the real balance effect according to Pigou, or, more likely, destabilizes the economy further via debt-deflation according to Fisher/Tobin/Minsky.) So it's not obvious how sluggish adjustment towards equilibrium must be for persistent unemployment to be a disequilibrium phenomenon (just as it's not obvious whether the observed frequency of price adjustment can explain monetary non-neutrality in a New Keynesian model, due to strategic complementarities) - indeed, if the equilibrium wasn't stable, no speed of price adjustment would be fast enough. To answer this question we would need a viable disequilibrium research program, which, sadly, doesn't exist (to my knowledge). I admit that the difficulties with constructing such a research program are a good argument for pursuing alternatives which do not stray too far from the equilibrium fold, e.g. incomplete labor market models.<br /> <br />(Even aside from its relevance to macro, I think the disequilibrium project is interesting for the same reason that microfoundations are usually interesting - it is more satisfying to know how and why prices and expectations adjust to bring people's plans into equilibrium, than to just assume they do so. General equilibrium theory lacks microfoundations. But this is a separate question.)Keshavnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-18218907159523864112012-04-19T20:18:19.247-04:002012-04-19T20:18:19.247-04:00I agree. The confusion is over the three uses of e...I agree. The confusion is over the three uses of equilibrium (see the post by Keshav).Roger Farmerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-60674107933407457242012-04-19T20:16:31.152-04:002012-04-19T20:16:31.152-04:00"Neoclassical methodology is basically a gian..."Neoclassical methodology is basically a giant tautology."<br /><br />So is the principle of action at a distance. But that doesn't make it useless. We need a language to organize our discourse and the DSGE language is an immensely useful way of communicating ideas.Roger Farmerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-4181004425270122692012-04-19T20:00:25.775-04:002012-04-19T20:00:25.775-04:00@Anonymous
You are incorrect to assert that mult...@Anonymous <br /><br />You are incorrect to assert that multiplicity of equilibria (or even indeterminacy of equilibria) is a local phenomenon. As a counter-example, consider any well defined general equilibrium model of money where there is typically a continuum of non-stationary equilibria converging to a steady state where money has no value. In models of the inflation tax, this indeterminate steady state is one where money has positive value. As an example, see my 1997 Macroeconomic Dynamics paper with Michael Woodford.Roger Farmerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-34367388053783985722012-04-19T19:51:03.833-04:002012-04-19T19:51:03.833-04:00There is a theorem in finite general equilibrium m...There is a theorem in finite general equilibrium models which asserts that under certain regularity conditions, the number of equilibria is finite and odd. Kehoe and Levine have an econometrica paper which shows that the same is not true in overlapping generations (OG) models and my work with Jess Benhabib showed that this result also breaks down once there are increasing-returns to scale in the technology. <br /><br />In OG models and in models with externalities, there may be a continuum of dynamic equilibria. The first-generation indeterminacy literature exploited that result to show that there are equilibria in which beliefs influence outcomes. Dave Cass and Karl Shell called this idea 'sunspots'.<br /><br />I think of incomplete labor market models as 'second generation' indeterminacy models. Here the problem is not that there are many dynamic equilibria; it is indeterminacy of the steady state.Roger Farmerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-87359901654255528722012-04-19T19:35:04.090-04:002012-04-19T19:35:04.090-04:00Noah I haven't read the Lo paper - but I will ...Noah I haven't read the Lo paper - but I will look at it. Rather than disequilibrium prices,I prefer to think of disequilibrium expectations. If the fundamentals are non-stationary and shifting then it would make sense for an agent to use a learning rule that puts more weight on recent information (constant gain learning). That would make sense, even in a unique equilibrium model. <br /><br />The reason I am not totally sold on that approach is that I do not see that it can easily explain a decade of 15% unemployment. I do not see a good culprit to explain which of the fundamentals shifted to cause the Great Depression. In my approach, the Depression was an equilibrium phenomenon. But not in the RBC sense of the term. It was an equilibrium with involuntary unemployment.Roger Farmerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-10685589951522581502012-04-19T17:23:08.391-04:002012-04-19T17:23:08.391-04:00Roger -
I haven't read the old literature on...Roger - <br /><br />I haven't read the old literature on non-Walrasian trades that you reference in your post. But here's my question: What happens when the equilibria shift on a time-scale similar to the time-scale at which prices change? Might the process by which prices converge to Walrasian values then be important in predicting which equilibrium the economy will move towards? It seems to me that it might. Have you read that Andrew Lo paper I linked to in my last post?Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-64175693119516798142012-04-19T17:13:05.297-04:002012-04-19T17:13:05.297-04:00@Keshav A number of responders have criticized me ...@Keshav A number of responders have criticized me for confounding Nash equilibrium with Walrasian equilibrium and the point is well taken. You are correct to point to three different uses of the equilibrium concept and I do not want to disparage a research agenda that seeks to investigate, more closely, the connections between (1) and (2). <br /><br />But I do not think that that agenda will help us to fix the problems with existing DSGE models in macro. Lucas pointed out that we cannot observe a market in disequilibrium. He was right. Indeed, in the sense in which the word is used in macroeconomics, we cannot observe a market. The best we can hope for is to observe sequences of trades and the prices at which those trades take place.<br /><br />How should we organize our observations? In my view, it is helpful to view agents as goal oriented decision makers who take actions to achieve what they perceive to be optimal outcomes. In a game theoretic formulation of an economy, we would need to specify the information structure of each agent, the timing of moves, the payoff structure of the game and the space of actions. That way of proceeding would explain data as a consistent set of plans; definition (2) in your taxonomy.<br /><br />The general equilibrium approach of Debreu Chapter 7 ( your definition (1)), takes a short cut. No single agent sets prices; but prices are chosen such that quantities, prices and expectations of future prices are mutually consistent. In this rational expectations equilibrium environment, the data are explained as a sequence of choices that obey the equilibrium consistency requirements at all points in time and in all states of nature.<br /><br />Is that a useful way to seek to understand the world? That depends on two things. First, does it help us to understand data? Second, does it provide us with a guide to good policy? If one sticks to standard DSGE models with a unique equilibrium ( in the sense of (1) in (2)), a case can be made that the answer to both questions is no. But the same charge cannot be leveled at DSGS models with incomplete labor markets.<br /><br />Explaining persistent unemployment as the stubborn failure of prices to adjust to their equilibrium values will go only so far. At some point we should entertain the idea that persistent unemployment is itself an equilibrium phenomenon, and at that point, the idea of trading at 'false prices', in other words 'disequilibrium', becomes an impediment to our understanding. <br /><br />I should add that, by multiple equilibrium models, I do not restrict myself to dynamical systems where there is more than one steady state. Incomplete labor market models have a continuum of steady state equilibria as well as a continuum of non-stationary equilibria, where equilibrium here is in your sense (1). When closed with a specification of how agents form beliefs, these models are capable of understanding economic data and of guiding economic policy.Rogerhttp://www.rogerfarmer.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-57268062017248920852012-04-19T16:42:49.981-04:002012-04-19T16:42:49.981-04:00No, I think you're right...No, I think you're right...Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-85227292258978475522012-04-19T16:42:14.331-04:002012-04-19T16:42:14.331-04:00Yes, I also don't understand Farmer's read...Yes, I also don't understand Farmer's readiness to dismiss non-Walrasian states of the world. But he knows more than I do, so maybe we should get him to elaborate!<br /><br />As for RE, does it really mean anything useful when your economy starts having sunspot-type behavior?Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-2093173784666445772012-04-19T14:34:51.547-04:002012-04-19T14:34:51.547-04:00The epistemology of these discussions breaks open ...The epistemology of these discussions breaks open the fragile eggshell of what we perceive to be the truth. Roger Farmer's research agenda won't change now; he has vested his career in his perception that general equilibrium --irrespective of whether it is Walrasian or the legerdemain of Arrow-Debreau-- is reasonable and as such we are confined to tread the path of believing (as Stephen Mark Cohn put it) in "a micro-founded auctioneer model of the macro economy". What's wrong with macro foundations where there is a continuous feedback between macro and micro levels? Agents (be they households, firms, etc.) neither make decisions in a vacuum nor do they have foresight to plan ahead --I am not referring to the perfect forecasting straw man argument-- they must deal with institutional realities of oligopoly, market power, emotion, class (oh, sorry - that is for Marxist analysis). <br />Time to put the Lucas critique into the recycle bin and hit empty and there is no point in hedging Freshwater with Saltwater. Here's to hoping for some superior thinking from the next generation.Anonymoushttps://www.blogger.com/profile/07857616719536009725noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-53073944277642749512012-04-19T14:11:23.449-04:002012-04-19T14:11:23.449-04:00The problem with equilibrium in current market is ...The problem with equilibrium in current market is fact, that behaving in not very reasonable way (speculations, feeding up bubble) is more profitable, than looking in to further horizon, and focusing on longer term profit – which is more “reasonable” for keeping the continuity. The lack of equilibrium is just result of wrong regulation and some kind of dysfunction of the capital market. We can call it fraud or speculation, it is not very important, the fact is that the biggest profit is taken from actions that put market out of equilibrium. Market in capitalism is free, that is obvious, but there are regulations and tax, which can shape profit levels for many different strategies. If free market was regulated against speculation and bubbles, it would be in equilibrium much of the time. The problem is in stabilization of the market. The rules and taxation need to be shaped in a way it maximizes profit for strategies, that are targeting at equilibrium. Market must be stable for the good of capitalism and economy. STABLE means that: for little (finite) disturbance at input there is little (finite) disturbance at output. Regulations and taxation should address that problem. Market is in dysfunction if for small change in input output is huge and goes to almost infinity (this is just bubble). In that condition, it just can't regulate itself, and it just doesn't work. STABILITY of market guarantees its proper functioning. Otherwise bailouts are needed, each and every time global leader are speaking the market is freaking out, bubbles are just everyday life, etc.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-46221641918964473502012-04-19T12:11:47.233-04:002012-04-19T12:11:47.233-04:00"Once one applies Debreu's vision of gene..."Once one applies Debreu's vision of general equilibrium theory to macroeconomics, disequilibrium becomes a misleading and irrelevant distraction."<br /><br />Also sorry, but this is an absolutely absurd statement and anyone fooled by it should take a course in basic logic. I might as well say:<br /><br />"Once one applies the Creationist vision of Intelligent Design to the analysis of biology evolution becomes a misleading and irrelevant distraction."<br /><br />In methodology we call such a statement a 'tautology'. Neoclassical methodology is basically a giant tautology. This is why many serious methodologists basically stopped engaging with the discipline in the late-70s.TheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-82994662881079556272012-04-19T12:09:13.594-04:002012-04-19T12:09:13.594-04:00Very interesting stuff for someone with a backgrou...Very interesting stuff for someone with a background in population ecology, another field in which we often use analogies to physical dynamical systems to think about what's going on.<br /><br />From that perspective, I found Roger Farmer's comments a little difficult to understand. It's quite common in ecology and evolutionary biology to write down and analyze dynamical models of, say, the population dynamics of interacting species (predator and prey for instance) in which those species adaptively adjust their behavior. Further, the details of how "adaptive" adjustment of behavior happens can be varied to reflect all kinds of things (e.g., the organisms only have limited information, etc.). Such systems have features like Nash equilibria that can be understood with game theoretic-type analyses, but they're still dynamical systems and so it still makes perfect sense to talk about their equilibria, local and global stability of those equilibria, other sorts of attractors like limit cycles or chaos, bifurcations, etc. Is the same not true in economics? Am I missing or misunderstanding something here?Jeremy Foxhttp://oikosjournal.wordpress.comnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-11715394116125858762012-04-19T12:01:34.976-04:002012-04-19T12:01:34.976-04:00"The idea of disequilibrium is borrowed from ..."The idea of disequilibrium is borrowed from the physical sciences where it has meaning in the context of, for example, Newtonian mechanics. A ball rolling down an inclined plane is an example of a physical system in disequilibrium. When it reaches the bottom of the plane, friction ensures that the ball will come to rest. That is an equilibrium. But it is not what we mean by an equilibrium in economics."<br /><br />This shows a complete misunderstanding of the meaning of the equilibrium metaphor in economics (yes folks, its a metaphor). As Farmer goes on to point out equilibrium is a state in which the economy is static. Same thing as the ball that has come to a standstill.<br /><br />Disequilibrium theory -- in both physics and economics -- has to do with movement and dynamics that have no ultimate 'end point'. Equilibrium theory, on the other hand, does posit an 'end point' (albeit usually in the future). This feeds into the fact that equilibrium theory is implicitly teleological (and outdated!) while disequilibrium theory is evolutionary or teleonomical (and not outdated!).<br /><br />All the other social sciences have dealt with these problems over the course of the last 50 years -- from psychology through anthropology to sociology. Economics remains stubbornly stuck at the point of not even recognising the metaphors they are deploying or the larger methodological issues. Some have -- like Mirowski -- but they are politely ignored.<br /><br />In saying this, I just got back from a conference yesterday where a neoclassical tried to model the modern banking system using systems dynamics. Afterwards I asked him what he thought such an approach might do to the idea of equilibrium. After a short conversation I'm fairly convinced he'll be moving far away from equilibrium analysis. <br /><br />Change will come as people apply new computer models to analysis of concrete problems. The profession haven't even begun to grasp what the equilibrium metaphor actually meant -- they could have figured it out, however, by reading Marx, Hegel, Whitehead or any other number of process philosophers.TheIllusionisthttps://www.blogger.com/profile/17642837989235595346noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-84028553930659761542012-04-19T10:43:49.720-04:002012-04-19T10:43:49.720-04:00This is probably a topic for a longer post but I h...This is probably a topic for a longer post but I have to (respectfully) disagree with Farmer here. Equilibrium requires the mutual consistency of individual plans. In Arrow-Debreu this is brought about by the coordinating role of prices, given the assumption that we have complete futures markets for all contingent claims. But RE assumes that individual plans are mutually consistent without concerning itself with the process by which such consistency is attained. In other words, without concerning itself with the stability of equilibria. Even when there is a unique equilibrium path, it need not be locally stable under a broad class of disequilibrium adjustment dynamics (see for instance Howitt, JPE 1992). In fact, I think that such instability is at the heart of economic crises. It's not multiplicity of equilibria that is the problem, it's the instability of equilibria with respect to disequilibrium dynamics. Explicit consideration of such dynamics is an important area of research which should be encouraged rather than shut down. <br /><br />Having said that, Farmer does make a very important point that RE is consistent will all kinds of economic fluctuations and crises. One could certainly model economies in this way. The question is whether one should.Rajivhttps://www.blogger.com/profile/13667685126282705505noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-16247142826092980672012-04-19T09:15:15.724-04:002012-04-19T09:15:15.724-04:00The point is that at some point the modeler assume...The point is that at some point the modeler assumes that the linearization is sufficient. At this point there is only one equilibrium. Also, this requires a model form where the model is linearizable in the first place--it is certainly possible to have a model form which is not linearizable.Zathrashttps://www.blogger.com/profile/14916717939276205773noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-29313112818577928382012-04-19T06:40:36.432-04:002012-04-19T06:40:36.432-04:00Well, if you prefer, the model tells you which are...Well, if you prefer, the model tells you which are the fundamental forces, how they interact, and what is the expected outcome (apple on the ground). It is indeed description rather than explanation. This is a semantic issue, it does not change the point.Zorbloghttps://www.blogger.com/profile/03175779746756984479noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-7831198349872982292012-04-19T06:13:27.627-04:002012-04-19T06:13:27.627-04:00I think Farmer is overselling his own work here.
...I think Farmer is overselling his own work here.<br /><br />First of all, in a model with a unique equilibrium we can easily have "sentiments" create huge (welfare disastrous) swings in aggregates. For instance, if agents think that with a certain probability the credit crunch will reduce investment opportunities and lower future growth, then they will start to save, spend less, and a crisis will emerge already today. Even if there has been no change to current fundamentals (only to the probability of future fundamentals). So what is the difference between this and sunspot models? Well, in sunspot models you can essentially say, "suppose I think I will behave like an erratic idiot tomorrow, then it is optimal to behave like an erratic idiot today", but since it is "optimal" being an "idiot" has no meaning, the belief is self-fulfilling. The "belief function" would guide us to how "idiotic" (or smart) we think of ourselves in the future, and limit the veritable plethora of retarded outcomes. I am not sure this is the way forward.<br /><br />Second, the multiplicity of equilibria is often a very local phenomenon, and only holds in linearized models. Once you leave that realm, models can instead be entirely explosive, or collapsing. Do we really think that it's a good idea to exploit knife-edge properties which only emerges under a silly approximation as the main explanation of crises. I doubt that.Anonymousnoreply@blogger.com